Utility of prevertebral soft tissue measurements in identifying patients with cervical spine fractures.

Study Objective: Prevertebral soft tissue measurements of more than 6 mm at C2 and more than 22 mm at C6 have been reported as radiologic evidence of cervical spine injury. The objective of this study was to determine the sensitivity and specificity of soft tissue measurements in patients with radiographically proven cervical spine fractures.
Design: Retrospective case control study.
Setting: Level I trauma center emergency department.
Participants: The study group consisted of patients admitted between January 1989 and August 1991 with an admitting or discharge diagnosis of cervical spine fracture. The control group was a systematic sampling of trauma patients seen in the ED during July 1991 who received a cervical spine radiograph. Patients less than 17 years old with penetrating injuries or injuries more than 24 hours old were excluded. One hundred thirty-eight study patients and 134 control patients were identified; 32 study patients and 41 control patients were excluded due to inaccessible records. One hundred six study patients and 93 control patients were used for data analysis.
Results: Study patients were divided into two groups: those with fractures at C1-C4 (n = 55) or C4-C7 (n = 86). A C2 prevertebral soft tissue measurement of more than 6 mm had a sensitivity of 59% and a specificity of 84% for fractures at C1-C4. A C6 prevertebral soft tissue measurement of more than 22 mm had a sensitivity of 5% and a specificity of 95% for fractures at C4-C7. Receiver operator characteristic curves for measurements at C2 and C6 failed to demonstrate a cutoff value with adequate sensitivity and specificity in detecting fracture.
Conclusion: We conclude that using prevertebral soft tissue measurements of more than 6 mm at C2 and more than 22 mm at C6 as a marker of cervical spine injury fails to identify a large proportion of patients with cervical spine fractures.